ORIGINAL ARTICLE

Lung function in poorly controlled type 1 North

African diabetic patients: A case-control study

Ines Slim a,b,1, Ferdaws Khalaf a,b,1, Imed Latiri c, Zouhour Elfkih a,b,

Sonia Rouatbi c,d, Ines Khochtali e, Ines Ghannouchi c,d, Abir Zinelabidine f,

Leila Ben Othman f, Hedi Miled g, Larbi Chaieb a,b,2, Helmi Ben Saad c,d,h,*,2

a Department of Endocrinology and Diabetology, Farhat HACHED University Hospital of Sousse, Tunisiab Endocrinology and Metabolic Diseases Unit, 02/UR/08-07, Faculty of Medicine of Sousse, University of Sousse, Sousse, Tunisiac Laboratory of Physiology, Faculty of Medicine of Sousse, University of Sousse, Sousse, Tunisiad Department of Physiology and Functional Exploration, Farhat HACHED University Hospital of Sousse, Tunisiae Department of Endocrinology and Diabetology, Fattouma BOURGUIBA University Hospital of Monastir, Tunisiaf Laboratory of Biochemistry, Basic Health Group, Sousse, Tunisiag Laboratory of Biochemistry, Farhat HACHED University Hospital of Sousse, Tunisiah Research Laboratory N� LR14ES05: Interactions of the Cardiopulmonary System, Faculty of Medicine of Sousse, University

of Sousse, Tunisia

Received 3 December 2014; accepted 15 February 2015

KEYWORDS

Endocrinology;

Diabetes mellitus;

Lung function tests;

Tunisia;

Abstract Aim: To compare the lung function parameters of poorly controlled type-1-diabetes-

mellitus (T1DM) patients with age-; height and sex-matched healthy-non-smokers (HNS).

Population and methods: Subjects aged 35–60 Yrs who have a poorly controlled T1DM

(glycated-Haemoglobin level >7%) with a disease history of more than 10 Yrs (n= 14) and

HNS subjects (n= 14) were recruited. Clinical, anthropometric and fasting biological data were

Abbreviations: ATS, American-thoracic-society; BMI, body-mass-index; CLA, chronological-lung-age; DLCO, capacity-to-transfer-carbon-

monoxide; DN4, douleur-neuropathique-4-questions; ELA, estimated-lung-age; ERS, European-respiratory-society; FEV1, first-second-forced-

expiratory-volume; FVC, forced-vital-capacity; HbA1c, glycated-haemoglobin; HDL-cholesterol, high-density-lipoprotein-cholesterol; HNS,

healthy-non-smokers; LDL-cholesterol, low-density-lipoprotein-cholesterol; LLN, lower-limit-of-normal; MMEF, maximal-mid-expiratory-flow;

NY, narghile-years; LAOVD, large-airways-obstructive-ventilatory-defect; PY, pack-years; RV, residual-volume; RVD, restrictive-ventilatory-

defect; SD, standard-deviation; SVC, slow-vital-capacity; T1DM, type-1-diabetes-mellitus; T2DM, type-2-diabetes-mellitus; TGV, thoracic-gas-

volume; TLC, total-lung-capacity; 95% CI, 95% confidence interval* Correspondent author at: Laboratory of Physiology, Faculty of Medicine of Sousse, Street Mohamed KAROUI, Sousse, Tunisia. Tel.: +216

98697024; fax: +216 73224899.

E-mail address: helmi.bensaad@rns.tn (H. Ben Saad).1 These authors contributed equally as first authors to this work.2 These authors contributed equally as senior authors to this work.

Peer review under responsibility of The Egyptian Society of Chest Diseases and Tuberculosis.

The French version of the present study abstract was accepted as POSTER DISCUSSION at the annual Congress of the French Society of

Pulmonology (SPLF, 31 January, 2 February 2015, Lille, France; http://www.sciencedirect.com/science/article/pii/S0761842514004112).

Name and location of the institution where the study was performed: service of Physiology and Functional Exploration and Department of

Endocrinology and Diabetology, Farhat HACHED Hospital. Sousse, Tunisia.

Egyptian Journal of Chest Diseases and Tuberculosis (2015) xxx, xxx–xxx

HO ST E D BY

The Egyptian Society of Chest Diseases and Tuberculosis

Egyptian Journal of Chest Diseases and Tuberculosis

www.elsevier.com/locate/ejcdtwww.sciencedirect.com

http://dx.doi.org/10.1016/j.ejcdt.2015.02.0130422-7638 ª 2015 The Authors. Production and hosting by Elsevier B.V. on behalf of The Egyptian Society of Chest Diseases and Tuberculosis.This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Please cite this article in press as: I. Slim et al., Lung function in poorly controlled type 1 North African diabetic patients: A case-control study, Egypt. J. Chest Dis.Tuberc. (2015), http://dx.doi.org/10.1016/j.ejcdt.2015.02.013

Aging;

Case-control studycollected. Plethysmographic data (flows, volumes, estimated-lung-age (ELA), lung-capacity-to-

transfer-carbon-monoxide (DLCO)) were measured. Large-airway-obstructive-ventilatory-defect

(LAOVD) was defined as first–second-forced-expiratory-volume (FEV1)/forced-vital-capacity

(FVC) below the lower-limit-of-normal (LLN). Restrictive-ventilatory-defect (RVD) was defined

as total-lung-capacity (TLC) < LLN. Lung-hyperinflation was defined as residual-volume

(RV) > upper-limit-of-normal. Student t-test and chi-2 test were used to compare plethysmo-

graphic data and profiles of the two groups.

Results: The two groups were matched in chronological-lung-age (CLA) (respectively 47 ± 7 vs.

50 ± 8 Yrs) and sex (7 males and 7 females in each group) and height. Compared to the HNS

group, the T1DM one had significantly lower FEV1, FVC, slow-vital-capacity and maximal-mid-

expiratory-flow (respectively 99 ± 11% vs. 83 ± 11%, 99 ± 9% vs. 86 ± 11%, 80 ± 8% vs.

67 ± 15% and 98 ± 23% vs. 72 ± 23%), had significantly higher TLC and RV (respectively,

105 ± 20% vs. 123 ± 24% and 108 ± 22% vs. 131 ± 24%) and had significantly higher percent-

age of subjects with lung-hyperinflation (7.1% vs. 43.0%). Both groups had similar percentages of

LAOVD and RVD and similar corrected DLCO values. ELA of the T1DM group (57 ± 10 Yrs)

was significantly higher than CLA.

Conclusion: Poorly controlled T1DM seems to alter ventilatory mechanics without effect on the

alveolo-capillary-membrane. In addition, it accelerates the respiratory ageing.

ª 2015 The Authors. Production and hosting by Elsevier B.V. on behalf of The Egyptian Society of Chest

Diseases and Tuberculosis. This is an open access article under the CC BY-NC-ND license (http://

creativecommons.org/licenses/by-nc-nd/4.0/).

Introduction

Diabetes-mellitus (DM), classified among the top ten leading

causes of death worldwide [1], is becoming a major public

health emergency [2]. In the Maghreb, a region undergoing

an epidemiological transition characterized by a decrease of

infectious diseases and an increase in chronic non-infectious

ones, the prevalence of DM was as high as 9.5% in 2012 [3].

There are twomajor types of DM: type 1 (T1DM) and type 2

(T2DM)); differentiated according to the aetiopathogenic fac-

tors and the rapidity of pancreatic beta cells of Langerhans islets

apoptosis [4]. Since the discovery of insulin, the outcome of

T1DM has completely changed. Treatment options and extra-

renal-epuration techniques are considered as a turning point

in the disease history leading to the improvement of its outcome

[5]. Since then, we are no longer worried about the direct mor-

tality of the disease as much about morbidity associated to

metabolic emergencies, and to micro- and macroangiopathy

[6] which, in turn, may have a negative impact on the function

of internal organs [7]. In fact, diabetic microangiopathy specifi-

cally affects the eyes (retinopathy), kidney (nephropathy) and

peripheral nervous system (neuropathy) [7]. Since it possesses

a very wide capillary network and a significant amount of con-

nective tissue, the lung could be a suitable ‘‘target organ’’ [8].

This was earlier suggested by Schuyler et al. [9] more than

40 Yrs ago. These authors [9] investigated lung function in 11

patients with T1DM and age-matched normal control subjects.

This pilot study was the first to report measurements of nearly

all the available tests of lung function, including lung elasticity,

capacity-to-transfer-carbon-monoxide (DLCO), absolute tho-

racic-gas-volumes (TGV), airflow resistance and maximal

forced-vital-capacity (FVC) tests [9]. As their subjects were life-

long non-smokers without allergies or lung disease, the T1DM

lung elastic recoil decrease was interpreted to reflect effects of

DM on lung elastic proteins [9]. That was hence the first results

in the literature suggesting that the lung may be a ‘‘target

organ’’ of T1DM. Since then, studies analysing T1DM

ventilatory mechanics and/or pulmonary exchanger are more

numerous with controversial conclusions: no effect [9–13],

abnormal data [14–27] with conflicting reports on the nature

of spirometric alterations: restrictive [14–17] and/or large-air-

ways obstructive [16,17]-ventilatory-defects (respectively,

RVD, and LAOVD) and/or decreased DLCO [14,15,18,

23,26], and normal spirometry data but decreased DLCO [27].

In addition, the majority of these studies [9–27] used limited

methodology: combination of the two types of DM [14,19–

22]; inclusion of T1DMpatients with a high difference in disease

duration [17]; inclusion of both controlled and uncontrolled

T1DM [21]; lack of control groups such as healthy-subjects

[17,19]; measurements of expiratory flows only [14,17,20,

21,24], without measurement of lung volumes and/or lung-hy-

perinflation; non-application [10,17,21,22,24] of the latest lung

function international recommendations [28–30], no report of

the applied lung function guidelines [14,19,20,23]; use of

unspecified spirometric norms [10,14,17,19,22–24] or of inap-

propriate spirometric definitions [14,17,19,23] (e.g. application

of a fixed thresholds of 70% or 80% as a lower-limit-of-normal

(LLN)) and no calculation, in case-control studies [10,14,20,22–

24], of the required sample size which is a statistically crucial

point [31]. A recent metaanalysis [32] concluded that when stud-

ied in the absence of overt pulmonary comorbidity, both T1DM

and T2DM were associated with a modestly impaired pul-

monary function in a restrictive pattern. The results were

irrespective of body-mass-index (BMI), smoking, DM dura-

tion, and glycated-haemoglobin (HbA1c) levels. In subanalyses,

the association seemed to be more pronounced in T2DM than

in T1DM [32].

Taking into account those conflicting conclusions about the

effect of T1DM on lung function and the remaining unan-

swered question about the reality of diabetic pulmonary alter-

ation that was recently risen [7], the present study aimed to

compare the lung function parameters (i.e. plethysmographic

and DLCO data measured according to recent lung

function test guidelines [28–30]), of a poorly controlled

T1DM group (HbA1c >7%) having a history of disease more

than 10 Yrs with those of age-, sex- and height matched

2 I. Slim et al.

Please cite this article in press as: I. Slim et al., Lung function in poorly controlled type 1 North African diabetic patients: A case-control study, Egypt. J. Chest Dis.Tuberc. (2015), http://dx.doi.org/10.1016/j.ejcdt.2015.02.013

healthy-non-smokers (HNS) . The null hypothesis is that there

will be no difference between lung function data mean values

in both groups.

Population and methods

Study design

It is a case-control study performed over a period of six-

months with collaboration between Departments of

Physiology and Functional Exploration, of Endocrinology

and Diabetes, Laboratories of Hematology and of

Biochemistry (University Hospital of Farhat HACHED) and

of Biochemistry (Basic Health Group of Sousse), Sousse,

Tunisia. The study approval was obtained from the

Hospital’s Ethics Committee. Written and informed consent

was asked from all study participants. All participants have

received a report of their explorations.

Sample size

The null hypothesis was H0:m1 = m2 and the alternative

hypothesis was Ha:m1 = m2 + d where d is the difference

between two means and n1 and n2 are the sample size for

two groups (T1DM and HNS) such that N = n1 + n2. The

total sample size was estimated using the following formula

[31]: N = ((r+ 1)(Za/2 + Z1�b)2r2)/rd2 where Za is the nor-

mal deviate at a level of significance (=2.58 for 1% level of sig-

nificance); Z1�b is the normal deviate at 1 � b% power with

b% of type II error (=1.28 at 90% statistical power); r equal

to n1/n2, is the ratio of sample size required for two groups

(r = 1 gives the sample size distribution as 1:1 for the two

groups). r and d are the pooled standard-deviation (SD) and

difference of first–second-forced-expiratory-volume (FEV1)

means of the two groups. These two values were obtained from

a previous study of similar hypothesis [23] where researcher

found the mean FEV1 (%) in the two groups were 95.19%

and 81.28% respectively and a common SD of 13.48%. The

total sample size for the study was 28 subjects (14 T1DM

and 14 HNS).

Study population

Patients’ group

Only poorly controlled T1DM patients aged 35–60 Yrs with a

history of T1DM of more than 10 Yrs were included. Non-in-

clusion criteria were: controlled T1DM (HbA1c 67%); T2DM,

recent infection or acute metabolic complication (occurred less

than one week before the lung function measurements), history

of asthma, allergic rhinitis, atopy or chronic obstructive pul-

monary disease and oral corticosteroid treatment within four

weeks prior to lung function test or bronchodilators use,

patient under dialysis. Researchers have verified the folders

and files of all patients with T1DM followed in the

Department of Endocrinology and Diabetes; and have con-

tacted them.

Control group

HNS aged 35–60 Yrs were recruited among Hospital workers

and the parents of medical school students. Informational

letters clarifying the aims of the study were put up at the

Hospital and the local Medical School. In addition, an article

announcing the need for recruitment of healthy subjects was

posted in a social network service (Facebook pages of the per-

sons implicated in the study). Non-inclusion criteria were: his-

tory of smoking, DM, confirmed cardiovascular or

pulmonary diseases, respiratory symptoms (e.g. chronic cough,

wheezing, dyspnoea Pstage two [33]), thoracic surgery, mental

disease and chronic medication use. The discovery of a

LAOVD and/or a RVD was an exclusion criterion.

Medical questionnaire

A medical questionnaire [34] was used to assess several sub-

jects’ characteristics (smoking, medical, surgical and gynae-

cologist-obstetrics histories, medication use, several T1DM

characteristics and dyspnoea [33]). Cigarette and narghile use

was evaluated, respectively, in pack-Yrs (PY) and narghile-

Yrs (NY) [35]. Two groups of patients who have successfully

weaned off smoking (ex-smokers) were defined [0. No, 1.

Yes]. T1DM duration (Yr) and specific complications

(retinopathy (no/yes), dialysis (no/yes)) were noted. Histories

of cardiovascular diseases were searched: angina, heart rhythm

disorder and arterial hypertension.

Physical examination

Sex and age (Yr) were noted. Height (±0.01 m) was measured

with a height gauge shoes removed, heels joined, and back

straight. Weight (±1 kg) was measured and the BMI (kg/m2)

was calculated. Waist circumference (m) was measured [36].

Neurological exam and ‘‘douleur neuropathique 4 questions’’

(DN4) score [37] were conducted in order to assess diabetic

neuropathy. Two groups were identified [0. No symptomatic

neuropathy (DN4 score <4); 1. Symptomatic neuropathy

(DN4 score P4)].

T1DM diagnosis, metabolic data and applied metabolic

definitions

The positivity of anti-pancreatic antibodies (antiGAD and/or

antiIA2) at the diagnosis of DM was the proof of the auto-im-

mune aetiology [38]. The following haematological data were

measured/calculated: numbers of blood cells (white (103/mm3),

red (106/mm3) and platelets (103/mm3)) and haemoglobin level

(g/dl). Anaemia was defined as haemoglobin level <12 g/dl in

female and <13 g/dl in male [39]. Two groups were defined [0.

No anaemia; 1. Anaemia]. Leukocytosis [40] was defined as

white-blood-cell count>11.103/mm3. Two groups were defined

[0. No leukocytosis; 1. Leukocytosis]. Fasting-glycaemia

(mmol/l), total-cholesterol (mmol/l) and high-density-lipopro-

tein-cholesterol (HDL-cholesterol, mmol/l) were quantified by

spectrophotometry. Triglycerides (mmol/l) were quantified by

the enzymatic-colorimetric method. Low-density-lipoprotein-c-

holesterol (LDL-cholesterol, mmol/l) was calculated [41].

HbA1c (%) was quantified on haemolysed total blood (turbidi-

metric inhibition immunoassay). Poorly controlled T1DM was

retained when HbA1c >7%. Diabetic nephropathy was

assessed by searching for increased urinary albumin (mg/24 h)

excretion using 24-h collections and serum creatinine (lmol/L)

Lung Function in Uncontrolled Type 1 Diabetes Mellitus 3

Please cite this article in press as: I. Slim et al., Lung function in poorly controlled type 1 North African diabetic patients: A case-control study, Egypt. J. Chest Dis.Tuberc. (2015), http://dx.doi.org/10.1016/j.ejcdt.2015.02.013

[42]. Two groups were defined [0. No diabetic nephropathy

(albuminuria <30 mg/24 h); 1. Diabetic nephropathy

(albuminuria P30 mg/24 h)].

Lung function measurements

Plethysmographic measurements were performed with a body

plethysmograph (ZAN 500 Body II, Mebgrerate GmbH,

Germany), carefully following international recommendations

[29,30]. The plethysmographic technique and the FVC

manoeuvre are extensively described in the Supplementary

data section. The following plethysmographic parameters were

measured/calculated and compared to local predicted values

[43]: FEV1 (L), FVC (L), maximal-mid-expiratory-flow

(MMEF, L/s), slow-vital-capacity (SVC, L), FEV1/FVC and

FEV1/SVC ratios (absolute values), TGV (L), residual-volume

(RV, L) and total-lung-capacity (TLC, L). The following def-

initions were applied: LAOVD: FEV1/SVC and/or FEV1/FVC

ratios <LLN [44]; RVD: TLC <LLN [44] and lung-hy-

perinflation: RV >upper-limit-of-normal [45]. DLCO was

measured by the single-breath method performed following

international recommendations [28]. Two DLCO measure-

ments in sitting position were performed on each subject.

Subjects were asked to assume the appropriate position five

minutes before the test with an interval of at least 15 min

between each DLCO measurement (the best value as a percent-

age of predicted values). The following parameters were mea-

sured/calculated and compared with European predicted

values [46]: DLCO (mmol/min/kPa; %), corrected DLCO

(DLCO predicted for haemoglobin [28] (mmol/min/kPa)). A

DLCO lower than the LLN was considered as abnormal

[28]. In order to evaluate respiratory aging, estimated-lung-

age (ELA, Yrs) was calculated [47].

Study conduct

First visit (day one, between 14 hrs. and 16 hrs.): presentation

of the subject to the Department of Endocrinology and

Diabetes; signature of consent; medical and DN4 question-

naires and anthropometric data. Second visit (day two,

between 9 a.m. and 11 a.m.): fasting blood and urine samples,

injection of usual dose of basal and/or prandial insulin (the

thighs site of injection was avoided); normal and standard

breakfast intake (including 200 cc of milk or equivalent in

carbohydrate intake and two 40-carbohydrates portions of

bread or equivalent); lung function measurements.

Data analysis

The Kolmogorov–Smirnov test was used to analyse variable

distribution. When the distribution was normal and the vari-

ances were equal, the results were expressed by their

means ± SD and 95% confidence interval (95%CI).

Otherwise, results were expressed by their medians (1st–3rd

quartiles). For plethysmographic data, a percentage of change

was calculated (=(T1DM value � HNS value)/T1DM value)

[24]. T-test and chi-2 test were used to compare, respectively,

quantitative data and percentages. T-test was used to compare

quantitative data of T1DM non-smokers vs. T1DM smokers,

of male T1DM vs. male HNS and of female T1DM vs. female

HNS. The associations between plethysmographic data

expressed in percentage predict and HbA1c or T1DM duration

were evaluated by Pearson’s product–moment correlation.

Wilcoxon-test was used to compare the ELA with the chrono-

logical-lung-age (CLA) of the each group. Analyses were car-

ried out using the Statistica statistical software (Statistica

Kernel version 6; Stat Software. France). Significance was set

at the 0.05 level.

Results

Descriptive data

Among the 62 examined subjects, only 28 (14 HNS and 14

poorly controlled T1DM) were retained for analysis.

The two group characteristics’ are presented in Tables 1

and 2. The main conclusions of these Tables are: (i) both

groups were age, sex and height matched; (ii) compared to

the HNS group, the poorly controlled T1DM one had signifi-

cantly lower weight, BMI and waist circumference. It also has

significantly higher percentages of subjects having history of

cardiovascular diseases or dyslipidemia or who were cigarette

smokers. (iii) Compared to the HNS group, the poorly con-

trolled T1DM one had a significantly higher fasting-glycaemia

level, a lower haemoglobin level, a higher percentage of anae-

mic subjects and a lower percentage of subjects with measured

dyslipidemia.

The DM duration mean ± SD (minimum–maximum) was

21 ± 8 (10–32) Yrs. Four patients have a cardiovascular dis-

ease (arterial hypertension (n= 2), angina with arterial hyper-

tension (n = 1) and heart rhythm disorder (n= 1)). Smoking

has been stopped since a mean ± SD duration of 9 ± 8 Yrs

in the ex-smokers group. The mean ± SD of cigarettes use

of the five male smokers included in the study was 30 ± 15

PY. Only one male patient was ex-narghile smoker (20 NY).

Analytical data

Lung function data and profiles of the two groups are pre-

sented in Table 3. Compared to the HNS group, the poorly

controlled T1DM one had significantly lower SVC, FVC,

FEV1, MMEF, DLCO (percentage changes were, respectively,

+16%, +13%, +16%, +27% and +16%) and had signifi-

cantly higher TLC, RV (percentage changes were, respectively,

�17% and �21%). However, there was no statistically signifi-

cant difference between both groups in corrected DLCO.

Compared to the HNS group, the poorly controlled T1DM

one had significantly higher percentages of subjects with

abnormal SVC, FEV1, MMEF and RV.

Table 4 shows the characteristics of the poorly controlled

T1DM group divided by smoking status. There was no signifi-

cant difference between metabolic data of the two groups and

compared with the T1DM non-smokers subgroup, the T1DM

ex-smokers group has significantly lower FVC (92 ± 8% vs.

79 ± 10%, respectively) and FEV1 (88 ± 8% vs. 75 ± 11%,

respectively).

Table 5 shows the comparison between T1DM and HNS

groups divided by sex. Compared to HNS females, those with

T1DM had significant decreases in FEV1 and FVC. Compared

4 I. Slim et al.

Please cite this article in press as: I. Slim et al., Lung function in poorly controlled type 1 North African diabetic patients: A case-control study, Egypt. J. Chest Dis.Tuberc. (2015), http://dx.doi.org/10.1016/j.ejcdt.2015.02.013

to HNS males, those with T1DM had significant decreases in

FEV1, FVC and SVC as well as a significant increase in RV.

No significant correlation between HbA1c or DM duration

and none of the plethysmographic data was found.

The ELA of the T1DM group was significantly higher than

the CLA, respectively, 56.70 ± 9.89 vs. 47.43 ± 7.19 Yrs

(p= 0.012). However, no statistical significant difference was

found between the HNS group ELA and CLA, respectively,

52.02 ± 7.18 vs. 47.49 ± 7.99 Yrs (p= 0.454).

Discussion

The present study shows impaired lung function parameters in

14 patients with poorly controlled T1DM compared to age-,

sex- and height matched controls. Therefore, the null hypothe-

sis, that there will be no difference between the two groups’

lung function data mean values’ was rejected. This impairment

was associated neither with the T1DMmean duration nor with

the HbA1c levels. In addition, T1DM seems accelerating the

respiratory ageing.

Discussion of the methodology

A Medline/PubMed research was conducted on 8 June 2014

and using the following three Medical Subject Headings

‘‘Diabetes Mellitus, Type 1’’ and ‘‘Respiratory Function

Tests’’ and ‘‘Adults’’. Among bibliographies of retrieved

articles that were manually reviewed, only international

studies published during the last 10 Yrs and all published

African studies were retained. Supplementary Tables 1E

and 2E describe, respectively, the five international and the

four African published studies [10,14,17,19–24]. The above

studies have controversial conclusions and used limited

methodology:

� Combination of the two types of DM [14,19–23]. This could

be a source of ‘‘confusion’’ because they have different

aetiopathogenies. Unlike T2DM during which the apopto-

sis occurs slowly and mainly due to glucotoxicity and lipo-

toxicity [4], T1DM is associated with a very fast and

irreversible apoptosis provoked by auto-immune factors

that are genetically programmed [48].

� Inclusion of T1DM patients with a high difference in disease

duration [17]. The DM duration varied from three (range:

0.5–13 Yrs) [17] to 15 ± 7 Yrs [10]. This could be a source

Table 1 Characteristics of type 1 diabetes-mellitus (T1DM)

and healthy-non-smoker (HNS) groups.

HNS

(n= 14)

T1DM

(n= 14)

Percentage

change

Anthropometric data

Male/female 7/7 7/7

Age (Yr) 49.84 ± 7.99 47.43 ± 7.19 +5

Height (m) 1.63 ± 0.10 1.66 ± 0.10 �2

Weight (kg) 80 ± 12 68 ± 12 +15*

Body-mass-index

(kg/m�2)

30.4 ± 4.0 25.0 ± 5.0 +18*

Waist circumference (m) 0.97 ± 0.09 0.88 ± 0.12 +9*

Clinical data

DN4 Not

reported

4 ± 3

Symptomatic

neuropathy

Yes 0 (0.0%) 8 (57.2%)*

No 14 (100.0%) 6 (42.8%)*

Diabetic

retinopathy

Yes 0 (0.0%) 8 (57.2%)*

No 14 (100.0%) 6 (42.8%)*

Cigarette use Yes 0 (0.0%) 5 (35.7%)*

No 14 (100.0%) 9 (64.2%)*

Narghile use Yes 0 (0.0%) 1 (7.1%)

No 14 (100.0%) 13 (92.9%)

DyspnoeaP stage

2

Yes 0 (0.0%) 1 (7.1%)

No 14 (100.0%) 13 (92.9%)

History of

Cardiovascular

diseases

Yes 0 (0.0%) 4 (28.5%)*

No 14 (100.0%) 10 (71.5%)*

Dyslipidemia Yes 0 (0.0%) 3 (21.5%)*

No 14 (100.0%) 11 (78.5%)*

Dysthyroidy Yes 0 (0.0%) 2 (14.3%)

No 14 (100.0%) 12 (85.7%)

Abdominal surgery Yes 2 (14.3%) 6 (42.8%)

No 12 (85.7%) 8 (57.2%)

DN4: neurological exam and douleur neuropathique 4 questions.

Percentage change = (T1DM value � HNS value)/HNS value.

Data are mean ± SD for anthropometric data and DN4 and

number (%) for others.* p<.05 (test-t, or chi-2): HNS vs. T1DM.

Table 2 Biological parameters of type 1 diabetes-mellitus

(T1DM) and healthy-non-smoker (HNS) groups.

HNS

(n= 14)

T1DM

(n= 14)

Percentage

change

Biological data

Fasting-glycaemia

(mmol/L)

5.39 ± 0.54 10.28 ± 5.60 �91*

HbA1c (%) Not reported 10.71 ± 1.45

Total-cholesterol

(mmol/L)

4.70 ± 0.55 4.47 ± 0.53 +5

Triglycerides

(mmol/L)

1.11 ± 0.36 0.96 ± 0.53 +14

HDL-C (mmol/L) 1.05 ± 0.37 1.29 ± 0.42 �23

LDL-C (mmol/L) 3.15 ± 0.50 2.75 ± 0.55 +13

White-blood-cells

(103/mm3)

6.63 ± 1.16 7.16 ± 2.16 �8

Red-blood-cells

(106/mm3)

4.59 ± 0.32 4.29 ± 0.47 +7

Platelets (103/mm3) 227 ± 44 249 ± 66 �10

Haemoglobin (g/dl) 13.5 ± 1.1 11.6 ± 1.7 +14*

Serum-creatinine

(lmol/L)

81.79 ± 14.27 91.43 ± 37.01 �12

Urinary-albumin

(mg/24 h)

NR 8.91 ± 8.17

Biological disorders

Anaemia Yes 2 (14.3%) 12 (85.7%)*

No 12 (85.7%) 2 (14.3%)*

Leukocytosis Yes 0 (0.0%) 1 (7.1%)

No 14 (100.0%) 13 (92.9%)

Diabetic

nephropathy

Yes NR 2 (14.3%)

No NR 12 (85.7%)

NR: not reported. Percentage change = (T1DM value � HNS

value)/HNS value. Data are mean ± SD for biological data and

number (%) for others.* p<.05 (test-t, or chi-2): HNS vs. T1DM.

Lung Function in Uncontrolled Type 1 Diabetes Mellitus 5

Please cite this article in press as: I. Slim et al., Lung function in poorly controlled type 1 North African diabetic patients: A case-control study, Egypt. J. Chest Dis.Tuberc. (2015), http://dx.doi.org/10.1016/j.ejcdt.2015.02.013

of misinterpretation, since it was suggested that the longer

the duration of T1DM is, the greater the lung function

impairment will be [24]. T1DM mean duration of the pre-

sent study (21 ± 8 Yrs) was higher than what was reported

in other studies (Tables 1E and 2E).

� Inclusion of both controlled and uncontrolled T1DM (respec-

tively, 40% and 60% in Hickson et al. [21] study). This

could be a source of confusion since it was demonstrated

that the uncontrolled DM group, when compared to con-

trolled one, has a significant decrease in lung function

[20]. Unlike the majority of the published studies [9–27],

the present study included only poorly controlled patients.

� Lack of the control group such as healthy-subjects [17,19].

Cross-sectional studies [14,24], while practical, do not pro-

vide a good basis for establishing causality.

� Measurements of expiratory flows only [14,17,20,21,24],

without measurement of lung volumes (useful parameters

for the diagnosis of RVD [44] and/or lung-hyperinflation

[45]) or DLCO [10,19,23] (useful for the evaluation of the

alveolar-capillary-membrane [28]) or lung-age. One positive

point for the present study, as done by others [10,19], was

the correction of DLCO according to haemoglobin levels

[28]. Lung age, a parameter not previously evaluated, has

Table 3 Lung function data and plethysmographic profiles of

type 1 diabetes-mellitus (T1DM) and healthy-non-smoker

(HNS) groups.

HNS

(n= 14)

T1DM

(n= 14)

Percentage

change

Plethysmographic and DLCO data (data are mean ± SD)

SVC (%) 80 ± 8 67 ± 15 +16*

FVC (%) 99 ± 9 86 ± 11 +13*

FEV1 (%) 99 ± 11 83 ± 11 +16*

FEV1/SVC (absolute

value)

0.8 ± 0.05 0.8 ± 0.10 +0

FEV1/FVC (absolute

value)

0.8 ± 0.03 0.8 ± 0.07 +0

MMEF (%) 98 ± 23 72 ± 23 +27*

TGV (%) 97 ± 7 94 ± 12 +3

TLC (%) 105 ± 20 123 ± 24 �17*

RV (%) 108 ± 22 131 ± 24 �21*

DLCO (%) 109 ± 21 92 ± 22 +16*

Corrected DLCO

(%)

113 ± 24 113 ± 42 +0

Lung function profiles (data are number (%))

HNS (n= 14) T1DM (n= 14) Probability

Plethysmographic data or DLCO lower than the lower-limit-of-

normal

FVC 0 (0.0%) 2 (14.3%)

SVC 4 (28.5%) 12 (85.7%) #

FEV1 0 (0.0%) 3 (21.5%) #

FEV1/FVC 0 (0.0%) 0 (0.0%)

FEV1/SVC 0 (0.0%) 0 (0.0%)

MMEF 0 (0.0%) 3 (21.5%) #

TLC 0 (0.0%) 1 (7.1%)

DLCO 1 (7.1%) 3 (21.5%)

Corrected DLCO 2 (14.3%) 1 (7.1%)

Plethysmographic data higher than the upper-lower-of-normal

RV 1 (7.1%) 6 (43.0%) #

For abbreviations, see list of abbreviations. %: percent of predicted

values. Percentage change = (HNS value � T1DM value)/HNS

value.* p< 0.05 (test-t): HNS vs. T1DM.# p< 0.05 (chi-2): HNS vs. T1DM.

Table 4 Characteristics of type 1 diabetes mellitus group

divided by smoking status.

Non-smokers

(n= 8)

Ex-smokers

(n= 6)

Percentage

change

Anthropometric data

Male/female 1/7 6/0

CLA (Yr) 46.29 ± 7.90 48.96 ± 6.51 �6

ELA (Yr) 53.63 ± 7.90� 59.56 ± 11.95� �11

Height (m) 1.61 ± 0.08 1.72 ± 0.09 �7*

Weight (kg) 69 ± 14 68 ± 10 +1

BMI (kg.m�2) 26.5 ± 5.72 23.0 ± 3.3 +13

Waist

circumference (m)

91 ± 14 84 ± 7 +8

Metabolic data

Fasting-glycaemia

(mmol/L)

10.68 ± 4.648 9.75 ± 7.13 +9

HbA1c (%) 11.14 ± 1.458 10.13 ± 1.35 +9

Total-cholesterol

(mmol/L)

4.64 ± 0.458 4.24 ± 0.56 +9

Triglycerides

(mmol/L)

1.03 ± 0.385 0.87 ± 0.72 +16

HDL-C (mmol/L) 1.39 ± 0.359 1.15 ± 0.48 +17

LDL-C (mmol/L) 2.78 ± 0.527 2.70 ± 0.63 +3

White-blood-cells

(103/mm3)

7.23 ± 2.78 7.06 ± 1.16 +2

Red-blood-cells

(106/mm3)

4.23 ± 0.41 4.37 ± 0.56 �3

Platelets

(103/mm3)

268 ± 64 223 ± 64 +17

Haemoglobin

(g/dl)

11.2 ± 0.9 12.2 ± 2.4 �9

Serum creatinin

(lmol/L)

88.13 ± 40.26 95.83 ± 35.38 �9

Urinary albumin

(mg/24 h)

8.91 ± 9.97 8.91 ± 5.88 +0

Plethysmographic and DLCO data

FEV1/SVC

(absolute value)

0.86 ± 0.11 0.79 ± 0.07 +8

FEV1/FVC

(absolute value)

0.82 ± 0.07 0.78 ± 0.08 +5

FVC (%) 92 ± 8 79 ± 10 +14*

FEV1 (%) 88 ± 8 75 ± 11 +15*

MMEF (%) 77 ± 20 64 ± 26 +17

SVC (%) 67 ± 20 67 ± 6 +0

TGV (%) 99 ± 12 87 ± 8 +12

TLC (%) 126 ± 26 119 ± 24 +6

RV (%) 140 ± 26 118 ± 16 +16

DLCO (%) 92 ± 13 93 ± 31 �1

Corrected DLCO

(%)

96 ± 19 136 ± 54 �42

For abbreviations, see list of abbreviations. %: percent of predicted

values. Percentage change = (ex-smokers value � non-smokers

value)/ex-smokers value. Data are mean ± SD for biological data

and number (%) for others.* p<.05 (test-t) non-smokers vs. ex-smokers.� p<.05 (Wilcoxon matched pairs test) for the same group.

6 I. Slim et al.

Please cite this article in press as: I. Slim et al., Lung function in poorly controlled type 1 North African diabetic patients: A case-control study, Egypt. J. Chest Dis.Tuberc. (2015), http://dx.doi.org/10.1016/j.ejcdt.2015.02.013

been shown to be an acceptable way to communicate spiro-

metric results for smokers and/or patients with respiratory

impairment [47].

� Non-application in some studies [10,17,21,22,24] of the latest

lung function international recommendations [28–30]. While

old recommendations (American-thoracic-society (ATS)-

1987 [49] or ATS-1995 [50]) were applied in some studies

[10,17,21,22,24], the present one predated recent guidelines

[28–30]. Surprisingly, the published African studies

[14,19,20,23] have not mentioned which spirometry guide-

lines were used.

� Use of unspecified spirometric norms [10,14,17,19,22–24].

This could lead to misinterpretation of spirometry data in

a significant proportion of subjects [43] since spirometry

results are often expressed as percentage of predicted values

derived from local reference equations [43].

� Use of inappropriate spirometric definitions (e.g. application

of a fixed thresholds of 70% or 80% as a LLN [10,14,17,19–

24]). This approach has been widely criticized and more

importantly, clinicians may have to review and revise pre-

vious diagnosis [51]. The present study applied the recent

international definitions based on a 95%CI [44].

Paradoxically, in some studies [10,20–22,24], the applied

spirometric definitions to diagnosis ventilatory-defects were

not reported.

� Use of old spirometry equipment [21]. Hickson et al. [21]

results were calculated using an equipment (dry rolling-seal

spirometers) that gave different results than those currently

recommended by the ATS/ERS [29]. However, lung

function testing equipment and procedures have been

progressively refined over the last 10 Yrs in line with

recommendations that have been regularly updated by the

ATS/ERS [28–30].

� No calculation, in some case-control studies [10,14,20,22–24],

of the required sample size. This could be a statistically cru-

cial point [31]. The present study calculated sample size of

T1DM (n = 14) which was higher than the sample size of

some studies (n = 8 [19]; n = 12 [10]), but was smaller than

the sample size of other ones (n = 20 [52]; n= 27 [23,24];

n= 30 [14]; n = 39 [17]).

Populations’ characteristics

The percentage of T1DM patients with symptomatic diabetic

neuropathy (57.2%, Table 1) was similar to that reported in

the literature (65.3% [53]). The mean age of the patients group

in the present study (47 ± 7 Yrs) was intermediate with those

reported in other studies: it varied from 14 ± 6 (range: 7–

28 Yrs) [17] to 54 ± 10 Yrs [22]. It has been proved that

Table 5 Comparison between the type 1 diabetes mellitus (T1DM) and healthy non-smoker (HNS) groups divided by sex.

Females Males

T1DM (n= 7) HNS (n= 7) T1DM (n = 7) HNS (n = 7)

Anthropometric data

CLA (Yr) 46.77 ± 8.40 49.37 ± 7.94 48.09 ± 6.37 50.30 ± 8.64

Height (m) 1.60 ± 0.08 1.54 ± 0.07 1.72 ± 0.08 1.71 ± 0.04

Weight (kg) 71 ± 14 76 ± 12 66 ± 10 84 ± 12*

BMI (kg.m�2) 27.6 ± 5.16 31.9 ± 2.83 22.4 ± 3.4 28.8 ± 4.5*

Waist circumference (m) 93.86 ± 12.60 94.7 ± 4.31 82.29 ± 8.26 99.43 ± 11.30*

Metabolic data

Fasting-glycaemia (mmol/L) 11.34 ± 4.59 5.40 ± 0.60* 9.21 ± 6.66 5.39 ± 0.52*

Total-cholesterol (mmol/L) 4.66 ± 0.49 4.86 ± 0.69 4.27 ± 0.52 4.54 ± 0.34

Triglycerides (mmol/L) 0.96 ± 0.36 1.13 ± 0.36 0.96 ± 0.70 1.09 ± 0.40

HDL-C (mmol/L) 1.42 ± 0.38 1.17 ± 0.38 1.15 ± 0.44 0.93 ± 0.35

LDL-C (mmol/L) 2.81 ± 0.57 3.18 ± 0.64 2.69 ± 0.57 3.13 ± 0.35

White-blood-cells (103/mm3) 7.30 ± 2.99 6.67 ± 1.35 7.02 ± 1.07 6.58 ± 1.03

Red-blood-cells (106/mm3) 4.16 ± 0.39 4.41 ± 0.32 4.42 ± 0.53 4.76 ± 0.21

Platelets (103/mm3) 267 ± 70 227 ± 46 231 ± 62 228 ± 47

Haemoglobin (g/dl) 11.0 ± 0.8 12.8 ± 1.0* 12 ± 2 14 ± 1*

Serum-creatinin (lmol/L) 90.86 ± 42.68 71.14 ± 6.52 92.00 ± 33.85 92.43 ± 11.59

Plethysmographic data

FEV1 (%) 87 ± 7 101 ± 9* 79 ± 13 98 ± 12*

FVC (%) 93 ± 8 103 ± 9* 80 ± 10 95 ± 9*

MMEF (%) 75 ± 21 96 ± 21 68 ± 26 99 ± 28

FEV1/SVC (absolute value) 0.85 ± 0.11 0.86 ± 0.05 0.81 ± 0.08 0.83 ± 0.04

FEV1/FVC (absolute value) 0.80 ± 0.05 0.83 ± 0.02 0.80 ± 0.10 0.83 ± 0.04

SVC (%) 66 ± 21 79 ± 8 68 ± 6 81 ± 9*

TGV (%) 121 ± 24 108 ± 26 125 ± 27 103 ± 13

RV (%) 139 ± 28 114 ± 25 122 ± 18 101 ± 18*

TLC (%) 99 ± 12 102 ± 7 89 ± 9 93 ± 3

DLCO (%) 91 ± 14 108 ± 28 94 ± 29 111 ± 13

Corrected DLCO (%) 91 ± 14 108 ± 28 135 ± 49 118 ± 20

For abbreviations, see list of abbreviations. %: percent of predicted values. Data are mean ± SD.* p<.05 (t-test): T1DM vs. HNS for each sex.

Lung Function in Uncontrolled Type 1 Diabetes Mellitus 7

Please cite this article in press as: I. Slim et al., Lung function in poorly controlled type 1 North African diabetic patients: A case-control study, Egypt. J. Chest Dis.Tuberc. (2015), http://dx.doi.org/10.1016/j.ejcdt.2015.02.013

T1DM [54] and lung function [55] are sex-dependent. For that

reason, and as done in some studies [17,19,21–23], the present

one was sex-matched.

Limits of the present study

One major limitation of the present study was the inclusion of

six males T1DM ex-smokers (43%). This inclusion makes the

T1DM sample’s profile questionable and the question that

may be raised here is: ‘‘should current- or ex-smokers be

included in the T1DM patient group?’’ In similar studies

(Table 3E) performed among T1DM [26,52], or T2DM [56–

58] or mixed-patients [59,60], 5% [26], 10% [58], 19% [56],

24% [59], 30% [52], 37% [57] and 50% [60] of DM patients

were current-smokers and 12.5% [58], 18.4% [59] and 33%

[56,60] were ex-smokers. In addition, and due to the high

prevalence of smoking in DM patients (22% to 33% of them

are smokers [61,62]), the inclusion of subjects with such risk

factors make the samples more representative especially of

poorly controlled patients. Moreover, comparisons between

female or male groups (T1DM vs. HNS) showed the same lung

function impairment (Tables 4 and 5). So, restricting the analy-

sis to T1DM females who had never smoked, did not alter the

present study findings. At least a recent study [63] have exam-

ined pulmonary function in smokers with a >10-PY with and

without diabetes. Authors have concluded that participants

with DM were observed to have reduced pulmonary function

after controlling for known risk factors such as smoking [63].

Finally, in a systematic review and meta-analysis investigating

pulmonary function in DM, metaregression analyses showed

that between-study heterogeneity was not explained by BMI,

smoking, diabetes duration, or HbA1c [32].

Additional discussion about the study design, recruitment

method, medical questionnaire and data analysis is detailed

in the Supplementary data section.

Discussion of the results

Effect of T1DM on lung flows and ratios

Results about the T1DM effect on peripheral flows (e.g.

MMEF) are controversial: unchanged [10,19,24] vs. decreased

[14,20,23] values. In the present study, the reduction of MMEF

by 27% (Table 3) can be considered as a marker of small air-

way defect. It is fair to assume that the small airways are a

potential target for T1DM and other specified studies are wel-

come. In a case-control study [10] (12 T1DM non-athletes vs.

12 healthy-non-athletes), no statistical differences between

FEV1 and FVC values were found (Table 1E). In another

study [19], the FEV1 of the 23 DM patients was qualified as

normal (>80%) (Table 2E). However, in line with the major-

ity of studies [14,20–24], the present one confirms that T1DM

reduces FEV1 and FVC. These results are a marker of a large

airway defect[64]: the FEV1 and/or FVC reduction by almost

0.300 L (13–16%) is higher than the spontaneous variability

observed in healthy subjects (of 0.183 L for FEV1 and

0.148 L for FVC [44]) and higher than the 0.200 L (12%)

threshold used in reversibility test interpretation [44]. Is

T1DM incriminated in the genesis of LAOVD? Data aiming

to respond to that question are still controversial: whereas

some studies [14,20,23] show a decrease in FEV1/FVC ratio,

the present study (Table 3), like others [10,24] (Tables 1E

and 2E), does not support this hypothesis and shows

unchanged FEV1/FVC ratio between T1DM and HNS groups.

In addition, FEV1/SVC was not affected by T1DM, as pre-

viously shown by Berriche et al. [19].

Effect of T1DM on lung volumes

Data about the effect of T1DM on lung volumes are contro-

versial (Tables 1E and 2E). As observed in the present study,

SVC was significantly decreased in some other case-control

studies [22,23]. It was, however, qualified as normal (>80%)

with Berriche et al. [19]. In practice, SVC is an important

parameter used to diagnose LAOVD [64] or ‘‘trapping phe-

nomena’’ [29]. Contrary to Masmoudi et al. [23] study, where

TLC was decreased showing a tendency to a RVD, the present

study objectifies an increase in TLC values showing a tendency

to lung-hyperinflation. Some other studies [10,19] found no

significant change in TLC (Tables 1E and 2E). One major

result of the present study was the significant increase of RV

in T1DM (Table 3). RV is an interesting parameter used in

practice to diagnose lung-hyperinflation [45] and the 0.400 L

(21%) RV increase is significantly higher than the recom-

mended threshold of 0.300 L (10%) applied in reversibility test

interpretation [45]. The present result was in opposition to the

findings of Berriche et al. [19] where DM patients RV was

lower than 80%.

Effect of T1DM on DLCO

The present study T1DM group DLCO was significantly

decreased (Table 3), which probably indicates a parenchyma

dysfunction. However, as DLCO changes with haemoglobin

levels and as the T1DM group has a statistically lower haemo-

globin level (Table 2) and includes a statistically higher per-

centage of anaemic subjects, specific adjustments for this

parameter (e.g. use of corrected DLCO) should always be

made to ensure appropriate interpretation [28]. In the present

study, as found by others [10], corrected DLCO values were

unchanged (Table 3). On the contrary, a significant decrease

in corrected DLCO was noticed in a previous study [23].

However the haemoglobin level was not reported [23]

(Table 2E).

Plethysmographic and DLCO profiles

No patient with T1DM had LAOVD (Table 3). This result,

different from the one found by Suresh et al. [17] (7.7% of

T1DM present a LAOVD), is explained by the divergence in

the applied definitions to retain the LAOVD diagnosis

(FEV1 <80% and FVC >80% or FEV1/FVC < 0.70 in

Suresh et al. study [17] vs. FEV1/FVC or FEV1/SVC< LLN

in the present one). In fact, applied in the present study,

Suresh et al. [17] definition gives a percentage of 21.4%. Half

of T1DM patients had lung-hyperinflation (Table 3). This

result, not previously reported, had a practical interest. It

was shown that lung-hyperinflation has important conse-

quences that surpass the framework of respiratory mechanics

[45]. Almost seven percent of T1DM patients had RVD

(Table 3), which is much lower than the 43.6% reported by

Suresh et al. [17]. This difference could be explained by the

divergence in the applied definitions to retain the RVD diagno-

sis (FVC <80% and FEV1/FVC>0.7 [17] vs. TLC < LLN in

8 I. Slim et al.

Please cite this article in press as: I. Slim et al., Lung function in poorly controlled type 1 North African diabetic patients: A case-control study, Egypt. J. Chest Dis.Tuberc. (2015), http://dx.doi.org/10.1016/j.ejcdt.2015.02.013

the present study). In fact, the application of Suresh et al. [17]

definition, in the present study, gives a percentage of 21.5%. In

Abd El-Azeem et al. [14] study, respiratory-defect was pre-

dominantly restrictive (no percentage was given). Applied in

the present study, the definition used by Abd El-Azeem et al.

[14] (normal FEV1/FVC and reduced DLCO) to diagnose

RVD, gives a percentage of 21.5%. Almost one fifth of

T1DM patients had decreased DLCO values (Table 3). This

percentage becomes 7.1%, when DLCO was adjusted for hae-

moglobin levels (Table 3). To the best of the authors’ knowl-

edge, no previous study has reported a percentage of T1DM

with abnormal DLCO.

Correlation between HbA1c or DM duration and

plethysmographic data

Like some studies [17,19,21] including T1DM patients with a

high HbA1c level, the present one, do not find a significant

correlation between HbA1c and lung function data. This result

confirms that HbA1c, a classical biological data often used to

monitor the development of T1DM, does not appear as an

indicator of respiratory deficiency. Lung function data, too,

does not appear to be influenced by disease duration

[14,17,19,21], which is confirmed by the present study, carried

out on patients with a mean duration of T1DM of 21 Yrs.

However, Meo et al. [24] have concluded that the years of dis-

ease showed a dose–response effect on lung function: the

longer the duration of disease, the greater the lung function

impairment.

T1DM and respiratory aging

One of the major results of the present study was that T1DM

accelerated lung ageing by 8.74 ± 8.63 Yrs. This phenomenon

seemed to be accelerated by smoking status. In fact, compared

to T1DM non-smokers, those smokers have a significant

acceleration of lung ageing (7.35 ± 7.94 vs. 10.61 ± 9.28).

This result, not previously described, could be used to encour-

age smoking cessation [65].

How T1DM alters lung function?

The aetiology of ‘‘diabetic pneumopathy’’ (i.e. a typical

histopathological and functional lung involvement) [25] is still

a matter of debate. The development of this long-term com-

plication could be explained by the biochemical alteration of

connective tissue proteins as well as microangiopathy of pul-

monary vessels [8,66], leading to mechanical lung dysfunction

and impaired blood gas exchange [67]. The mechanisms by

which T1DM could alter lung function are detailed in the

Supplementary data section.

Practical recommendations

It is advisable that patients with T1DM, mainly the poorly

controlled ones with duration of disease longer than 10 Yrs,

should undergo periodic plethysmographic tests (e.g. one

test/year) to assess the severity of lung function impairment

[24]. Plethysmography will identify more susceptible patients

with T1DM; so that they will be able to take additional pre-

ventive measures towards lung damage [24]. These measures

will help to prevent lung damage at the initial stage, which

often, over in the long run, contributes to morbidity and mor-

tality in diabetic patients [24]. Additionally, physicians should

contemplate the lung in the same way as any other com-

plications of T1DM. They should also recognize the size of

the problem of pulmonary complications as a consequence

of the novel techniques used through the respiratory tract in

the treatment of DM such as insulin pump inhaler [68].

Finally, in order to motivate T1DM smokers to stop smoking,

it is very useful to convey to them data about their ELA

[47,65].

Sources of financial support

None.

Authors’ contributions

IS conceived of the study, and participated in its design, and

performed the statistical analysis and coordination and

helped to draft the manuscript.

FK conceived of the study, participated in its design and

performed medical questionnaire and lung function tests,

and performed the statistical analysis.

IL conceived of the study, participated in its design and per-

formed lung function tests.

ZE conceived of the study, participated in its design and

performed lung function tests.

SR helped to draft the manuscript.

IK helped to draft the manuscript.

IG performed the lung function tests.

AZ realized and interpreted the biological analysis.

LBO realized and interpreted the biological analysis.

HM realized and interpreted the biological analysis.

LC helped to draft the manuscript and approved the final

version of the paper.

BSH conceived of the study, and participated in its design,

and performed the statistical analysis and coordination and

helped to draft the manuscript.

All authors read and approved the final manuscript.

Conflict of interest

The authors declare that they have no conflicts of interest con-

cerning this article.

Appendix A. Supplementary data

Supplementary data associated with this article can be found,

in the online version, at http://dx.doi.org/10.1016/j.ejcdt.2015.

02.013.

References

[1] C.J. Murray, A.D. Lopez, Measuring the global burden of

disease, N. Engl. J. Med. 369 (2013) 448–457.

[2] R. Bouguerra, H. Alberti, L.B. Salem, C.B. Rayana, J.E. Atti, S.

Gaigi, et al, The global diabetes pandemic: the tunisian

experience, Eur. J. Clin. Nutr. 61 (2007) 160–165.

[3] S. Hammami, S. Mehri, S. Hajem, N. Koubaa, H. Souid, M.

Hammami, Prevalence of diabetes mellitus among non

Lung Function in Uncontrolled Type 1 Diabetes Mellitus 9

Please cite this article in press as: I. Slim et al., Lung function in poorly controlled type 1 North African diabetic patients: A case-control study, Egypt. J. Chest Dis.Tuberc. (2015), http://dx.doi.org/10.1016/j.ejcdt.2015.02.013

institutionalized elderly in Monastir city, BMC Endocr. Dis. 12

(2012) 15.

[4] H.E. Thomas, M.D. McKenzie, E. Angstetra, P.D. Campbell,

T.W. Kay, Beta cell apoptosis in diabetes, Apoptosis 14 (2009)

1389–1404.

[5] R.G. Miller, A.M. Secrest, R.K. Sharma, T.J. Songer, T.J.

Orchard, Improvements in the life expectancy of type 1 diabetes:

the Pittsburgh epidemiology of diabetes complications study

cohort, Diabetes 61 (2012) 2987–2992.

[6] D. Daneman, Type 1 diabetes, Lancet 367 (2006) 847–858.

[7] K. Kuziemski, K. Specjalski, E. Jassem, Diabetic pulmonary

microangiopathy – fact or fiction?, Endokrynol Pol. 62 (2011)

171–176.

[8] F. Innocenti, A. Fabbri, R. Anichini, S. Tuci, G. Pettina, F.

Vannucci, et al, Indications of reduced pulmonary function in

type 1 (insulin-dependent) diabetes mellitus, Diabetes Res. Clin.

Pract. 25 (1994) 161–168.

[9] M.R. Schuyler, D.E. Niewoehner, S.R. Inkley, R. Kohn,

Abnormal lung elasticity in juvenile diabetes mellitus, Am.

Rev. Respir. Dis. 113 (1976) 37–41.

[10] W. Komatsu, T. Barros Neto, A. Chacra, S. Dib, Aerobic

exercise capacity and pulmonary function in athletes with and

without type 1 diabetes, Diab. Care 33 (2010) 2555–2557.

[11] G. Schernthaner, P. Haber, F. Kummer, H. Ludwig, Lung

elasticity in juvenile-onset diabetes mellitus, Am. Rev. Respir.

Dis. 116 (1977) 544–546.

[12] K. Strojek, D. Ziora, J.W. Sroczynski, K. Oklek, Pulmonary

complications of type 1 (insulin-dependent) diabetic patients,

Diabetologia 35 (1992) 1173–1176.

[13] A. Verrotti, F. Chiarelli, M. Verini, G. Morgese, Pulmonary

function in type 1 (insulin-dependent) diabetes mellitus,

Diabetologia 36 (1993) 579–580.

[14] I. Abd El-Azeem, G. Hamdy, M. Amin, A. Rashad, Pulmonary

function changes in diabetic lung, Egypt. J. Chest. Dis. Tuberc.

62 (2013) 513–517.

[15] M.S. Boulbou, K.I. Gourgoulianis, V.K. Klisiaris, T.S.

Tsikrikas, N.E. Stathakis, P.A. Molyvdas, Diabetes mellitus

and lung function, Med. Princ. Pract. 12 (2003) 87–91.

[16] P. Makkar, M. Gandhi, R.P. Agrawal, M. Sabir, R.P. Kothari,

Ventilatory pulmonary function tests in type 1 diabetes mellitus,

J. Assoc. Physicians India 48 (2000) 962–966.

[17] V. Suresh, A. Reddy, A. Mohan, G. Rajgopal, P. Satish, C.

Harinarayan, et al, High prevalence of spirometric

abnormalities in patients with type 1 diabetes mellitus, Pediatr.

Endocrinol. Diabetes Metab. 17 (2011) 71–75.

[18] D. Bell, A. Collier, D.M. Matthews, E.J. Cooksey, G.J.

McHardy, B.F. Clarke, Are reduced lung volumes in IDDM

due to defect in connective tissue?, Diabetes 37 (1988) 829–

831

[19] O. Berriche, F. Ben Mami, S. Mhiri, A. Achour, Is the

respiratory function altered during diabetes mellitus?, Tunis

Med. 87 (2009) 499–504.

[20] M.M. El-Habashy, M.A. Agha, H.A. El-Basuni, Impact of

diabetes mellitus and its control on pulmonary functions and

cardiopulmonary exercise tests, Egypt. J. Chest Dis. Tuberc. 63

(2014) 471–476.

[21] D.A. Hickson, C.M. Burchfiel, J. Liu, M.F. Petrini, K.

Harrison, W.B. White, et al, Diabetes, impaired glucose

tolerance, and metabolic biomarkers in individuals with

normal glucose tolerance are inversely associated with lung

function: the jackson heart study, Lung 189 (2011) 311–321.

[22] M. Irfan, A. Jabbar, A.S. Haque, S. Awan, S.F. Hussain,

Pulmonary functions in patients with diabetes mellitus, Lung

India 28 (2011) 89–92.

[23] K. Masmoudi, F. Choyakh, N. Zouari, Ventilatory mechanics

and alveolo-capillary diffusion in diabetes, Tunis. Med. 80

(2002) 524–530.

[24] S.A. Meo, A.M. Al-Drees, S.F. Shah, M. Arif, K. Al-Rubean,

Lung function in type 1 Saudi diabetic patients, Saudi Med. J. 26

(2005) 1728–1733.

[25] R.A. Primhak, G. Whincup, J.N. Tsanakas, R.D. Milner,

Reduced vital capacity in insulin-dependent diabetes, Diabetes

36 (1987) 324–326.

[26] C. Schnack, A. Festa, A. Schwarzmaier-D’Assie, P. Haber, G.

Schernthaner, Pulmonary dysfunction in type 1 diabetes in

relation to metabolic long-term control and to incipient diabetic

nephropathy, Nephron 74 (1996) 395–400.

[27] M.P. Villa, M. Montesano, M. Barreto, J. Pagani, M. Stegagno,

G. Multari, et al, Diffusing capacity for carbon monoxide in

children with type 1 diabetes, Diabetologia 47 (2004) 1931–1935.

[28] N. Macintyre, R.O. Crapo, G. Viegi, D.C. Johnson, C.P. van

der Grinten, V. Brusasco, et al, Standardisation of the single-

breath determination of carbon monoxide uptake in the lung,

Eur. Respir. J. 26 (2005) 720–735.

[29] M.R. Miller, J. Hankinson, V. Brusasco, F. Burgos, R.

Casaburi, A. Coates, et al, Standardisation of spirometry,

Eur. Respir. J. 26 (2005) 319–338.

[30] J. Wanger, J.L. Clausen, A. Coates, O.F. Pedersen, V. Brusasco,

F. Burgos, et al, Standardisation of the measurement of lung

volumes, Eur. Respir. J. 26 (2005) 511–522.

[31] K. Suresh, S. Chandrashekara, Sample size estimation and

power analysis for clinical research studies, J. Hum. Reprod. Sci.

5 (2012) 7–13.

[32] B. van den Borst, H.R. Gosker, M.P. Zeegers, A.M. Schols,

Pulmonary function in diabetes: a metaanalysis, Chest 138

(2010) 393–406.

[33] D.A. Mahler, C.K. Wells, Evaluation of clinical methods for

rating dyspnea, Chest 93 (1988) 580–586.

[34] B. Ferris, Epidemiology standardization project ii:

recommended respiratory disease questionnaires for use with

adults and children in epidemiological research, Am. Rev.

Respir. Dis. 118 (1978) 7–52.

[35] H. Ben Saad, The narghile and its effects on health. Part i: the

narghile, general description and properties, Rev. Pneumol.

Clin. 65 (2009) 369–375.

[36] R. Bouguerra, H. Alberti, H. Smida, L.B. Salem, C.B. Rayana,

J. El Atti, et al, Waist circumference cut-off points for

identification of abdominal obesity among the tunisian adult

population, Diab. Obes. Metab. 9 (2007) 859–868.

[37] G. Harifi, I. Ouilki, I. El Bouchti, M.A. Ouazar, A. Belkhou, R.

Younsi, et al, Validity and reliability of the arabic adapted

version of the dn4 questionnaire (douleur neuropathique 4

questions) for differential diagnosis of pain syndromes with a

neuropathic or somatic component, Pain Pract. 11 (2011) 139–

147.

[38] American Diabetes Association, Diagnosis and classification

of diabetes mellitus, Diabetes Care 36 (Suppl 1) (2013) S67–

S74.

[39] E. Villar, M. Lievre, M. Kessler, V. Lemaitre, E. Alamartine, M.

Rodier, et al, Anemia normalization in patients with type 2

diabetes and chronic kidney disease: results of the nephrodiab2

randomized trial, J. Diabetes Complications 25 (2011) 237–

243.

[40] N. Abramson, B. Melton, Leukocytosis: basics of clinical

assessment, Am. Fam. Physician 62 (2000) 2053–2060.

[41] M. Okada, H. Matsui, Y. Ito, A. Fujiwara, K. Inano, Low-

density lipoprotein cholesterol can be chemically measured: a

new superior method, J. Lab. Clin. Med. 132 (1998) 195–201.

[42] H. Kramer, M.E. Molitch, Screening for kidney disease in adults

with diabetes, Diabetes Care 28 (2005) 1813–1816.

[43] H. Ben Saad, M.N. El Attar, K. Hadj Mabrouk, A.B.

Abdelaziz, A. Abdelghani, M. Bousarssar, et al, The recent

multi-ethnic global lung initiative 2012 (GLI) reference values

don’t reflect contemporary adult’s north african spirometry,

Respir. Med. 2013 (107) (2012) 2000–2008.

10 I. Slim et al.

Please cite this article in press as: I. Slim et al., Lung function in poorly controlled type 1 North African diabetic patients: A case-control study, Egypt. J. Chest Dis.Tuberc. (2015), http://dx.doi.org/10.1016/j.ejcdt.2015.02.013

[44] R. Pellegrino, G. Viegi, V. Brusasco, R.O. Crapo, F. Burgos, R.

Casaburi, et al, Interpretative strategies for lung function tests,

Eur. Respir. J. 26 (2005) 948–968.

[45] H. Ben Saad, L. Ben Amor, S. Ben Mdalla, I. Ghannouchi, M.

Ben Essghair, R. Sfaxi, et al, The importance of lung volumes in

the investigation of heavy smokers, Rev. Mal. Respir. 31 (2014)

29–40.

[46] J.E. Cotes, D.J. Chinn, P.H. Quanjer, J. Roca, J.C. Yernault,

Standardization of the measurement of transfer factor (diffusing

capacity). Work group on standardization of respiratory

function tests. European community for coal and steel. Official

position of the european respiratory society, Rev. Malad.

Respir. 11 (Suppl 3) (1994) 41–52.

[47] H. Ben Saad, H. Selmi, K. Hadj Mabrouk, I. Gargouri, A.

Nouira, H. Said Latiri, et al, Spirometric ‘‘lung age’’ estimation

for north african population, Egypt. J. Chest Dis. Tuberc. 63

(2014) 491–493.

[48] D. Mathis, L. Vence, C. Benoist, Beta-cell death during

progression to diabetes, Nature 414 (2001) 792–798.

[49] Standardization of spirometry-1987 update. Statement of the

American thoracic society, Am. Rev. Respir. Dis. 136 (1987)

1285–98.

[50] American thoracic society, Single-breath carbon monoxide

diffusing capacity (transfer factor). Recommendations for a

standard technique–1995 update, Am. J. Respir. Crit. Care Med.

152 (1995) 2185–2198.

[51] Le groupe Pulmonaria, P.H. Quanjer, P.L. Enright, J. Stocks,

G. Ruppel, M.P. Swanney, et al, Open letter to the members of

the gold committee, Rev. Mal. Respir. 27 (2010) 1003–1007.

[52] L. Fuso, P. Cotroneo, S. Basso, M. De Rosa, A. Manto, G.

Ghirlanda, et al, Postural variations of pulmonary diffusing

capacity in insulin-dependent diabetes mellitus, Chest 110 (1996)

1009–1013.

[53] M. Halawa, A. Karawagh, A. Zeidan, A. Mahmoud, M. Sakr,

A. Hegazy, Prevalence of painful diabetic peripheral neuropathy

among patients suffering from diabetes mellitus in Saudi Arabia,

Curr. Med. Res. Opin. 26 (2010) 337–343.

[54] D.M. Maahs, N.A. West, J.M. Lawrence, E.J. Mayer-Davis,

Epidemiology of type 1 diabetes, Endocrinol. Metab. Clin.

North Am. 2010 (39) (2010) 481–497.

[55] M.R. Becklake, F. Kauffmann, Gender differences in airway

behaviour over the human life span, Thorax 54 (1999) 1119–

1138.

[56] H.C. Yeh, N.M. Punjabi, N.Y. Wang, J.S. Pankow, B.B.

Duncan, C.E. Cox, et al, Cross-sectional and prospective study

of lung function in adults with type 2 diabetes: the

Atherosclerosis Risk in Communities (ARIC) study, Diabetes

Care 31 (2008) 741–746.

[57] A.R. Ortiz-Aguirre, M.H. Vargas, A. Torres-Cruz, M. Quijano-

Torres, Age related spirometric changes in diabetic patients,

Rev. Invest Clin. 58 (2006) 109–118.

[58] A.C. Lau, M.K. Lo, G.T. Leung, F.P. Choi, L.Y. Yam, K.

Wasserman, Altered exercise gas exchange as related to

microalbuminuria in type 2 diabetic patients, Chest 125 (2004)

1292–1298.

[59] T.M. McKeever, P.J. Weston, R. Hubbard, A. Fogarty, Lung

function and glucose metabolism: an analysis of data from the

Third National Health and Nutrition Examination Survey, Am.

J. Epidemiol. 161 (2005) 546–556.

[60] P. Lange, S. Groth, J. Kastrup, J. Mortensen, M. Appleyard, J.

Nyboe, et al, Diabetes mellitus, plasma glucose and lung

function in a cross-sectional population study, Eur. Respir. J.

2 (1989) 14–19.

[61] E.S. Ford, A.M. Malarcher, W.H. Herman, R.E. Aubert,

Diabetes mellitus and cigarette smoking. Findings from the

National health interview survey, Diabetes Care 1994 (17) (1989)

688–692.

[62] R.I. Dierkx, W. van de Hoek, J.B. Hoekstra, D.W. Erkelens,

Smoking and diabetes mellitus, Neth. J. Med. 48 (1996) 150–

162.

[63] G.L. Kinney, J.L. Black-Shinn, E.S. Wan, B. Make, E. Regan,

S. Lutz, et al, Pulmonary function reduction in diabetes with

and without chronic obstructive pulmonary disease, Diabetes

Care 37 (2014) 389–395.

[64] H. Ben Saad, R. Ben Attia Saafi, S. Rouatbi, S. Ben Mdella, A.

Garrouche, A. Zbidi, et al, Which definition to use when

defining airflow obstruction?, Rev Mal. Respir. 24 (2007) 323–

330.

[65] S. Ben Mdalla, H. Ben Saad, N. Ben Mansour, B. Rouatbi, M.

Ben Esseghair, S. Mezghani, et al, The announcement of the

lung age it is a motivation to quit smoking?, Tunis Med. 91

(2013) 521–526.

[66] D.C. Weir, P.E. Jennings, M.S. Hendy, A.H. Barnett, P.S.

Burge, Transfer factor for carbon monoxide in patients with

diabetes with and without microangiopathy, Thorax 43 (1988)

725–726.

[67] M. Sandler, Is the lung a ‘target organ’ in diabetes mellitus?,

Arch Intern. Med. 150 (1990) 1385–1388.

[68] V.P. Guntur, R. Dhand, Inhaled insulin: extending the

horizons of inhalation therapy, Respir. Care 52 (2007) 911–922.

Lung Function in Uncontrolled Type 1 Diabetes Mellitus 11

Please cite this article in press as: I. Slim et al., Lung function in poorly controlled type 1 North African diabetic patients: A case-control study, Egypt. J. Chest Dis.Tuberc. (2015), http://dx.doi.org/10.1016/j.ejcdt.2015.02.013